On 1/31/08, Christian Heimes <[EMAIL PROTECTED]> wrote: > > Jesus Cea wrote: > > My guess is that 2.5 branch is still open to more patches than pure > > security/stability patches, so "backporting" BerkeleyDB 4.6 support > > seems reasonable (to me). If I'm wrong, please educate me :-). > > I think you are wrong, sorry pal! DB 4.6 support is a new feature. New > features must land in the development version(s) of Python, that is > Python 2.6 and 3.0. You must change as less code as possible in Python > 2.5 to fix a severe problem. > > Christian >
Actually in all past releaseXX-maint branches I have merged in support for compiling against a new version of BerkeleyDB. Its not a feature, its just something you have to do to support compiling against a new library version. Why is this ok? * There are no API changes on the python module side. * Binary releases of python for windows continue to be compiled against the same BerkeleyDB version that was used in the 2.5.0 release. the combo of those two things keeps it sane to do this. As Martin pointed out, I already merged that trivial change in to release25-maint a while back. That said, BerkeleyDB 4.6.x has proven to be a bit of an unstable release from Oracle so I'm considering modifying setup.py to disable linking against that version by default. I'll be reviewing buildbot results this weekend. As already mentioned by Neal i've disabled 4.6 support in trunk to watch the buildbots. I'll go over the results this weekend to make a decision on whether or not python's setup.py should default to allowing itself to link against 4.6.x or not in hopes of avoiding people filing bugs against us for what is ultimately Oracle's problem. -Greg
_______________________________________________ Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com
